© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 86136
    Prev
    Next
    INDEX
    Explanations

    scuba diving

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron selectively detects the token “scuba,” i.e. mentions of scuba diving.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     et
    -3.13
    é
    -2.77
    si
    -2.56
    '
    -2.48
    ’
    -2.44
    l
    -2.39
    "
    -2.27
     da
    -2.25
     of
    -2.20
     และ
    -2.16
    POSITIVE LOGITS
    because
    2.92
     носить
    2.27
    焿
    2.25
    they
    2.25
    ጴ
    2.23
    讣
    2.22
     ставить
    2.17
     искать
    2.17
    媄
    2.17
     插图
    2.16
    Activations Density 0.003%

    No Known Activations